Details
| Original language | English |
|---|---|
| Article number | 110159 |
| Journal | Mechanical Systems and Signal Processing |
| Volume | 191 |
| Early online date | 3 Feb 2023 |
| Publication status | Published - 15 May 2023 |
Abstract
Cross-domain machinery fault diagnosis aims to transfer enriched diagnosis knowledge from a labeled source domain to a new unlabeled target domain. Most existing methods assume that the prior information on the fault modes of the target domain is known in advance. However, in engineering practice, prior knowledge of fault modes is rare in a new domain, in which there may be only partial source fault modes or some new fault modes. Furthermore, up to the present, almost all existing cross-domain fault diagnosis methods require the labeled source data during the model training process, which restricts their deployment on certain devices with limited computing resources. To this end, we propose a universal source-free domain adaptation method that can handle cross-domain fault diagnosis scenarios without access to the source data and is free of explicit assumptions about the target fault modes. More specifically, we develop a convolutional network with a Transformer as the attention module to extract discriminative feature information from the source data and then send the model and parameters to the target domain. In target domain training, we first propose a supervised contrastive learning strategy based on source class prototypes, which utilizes high-confident predictions to achieve source-free domain alignment and class alignment. Then, we also introduce a threshold-based entropy max–min loss to further align known class samples in the target domain or reject target outlier samples as an unknown class. Furthermore, we introduce self-supervised learning to further learn feature representations of the target domain to reduce the previous misclassification. A series of experiments on two rotating machine datasets demonstrate the effectiveness and practicability of the proposed method.
Keywords
- Domain adaptation, Fault diagnosis, Machinery, Source-free, Supervised contrastive learning
ASJC Scopus subject areas
- Engineering(all)
- Control and Systems Engineering
- Computer Science(all)
- Signal Processing
- Engineering(all)
- Civil and Structural Engineering
- Engineering(all)
- Aerospace Engineering
- Engineering(all)
- Mechanical Engineering
- Computer Science(all)
- Computer Science Applications
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Mechanical Systems and Signal Processing, Vol. 191, 110159, 15.05.2023.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Universal source-free domain adaptation method for cross-domain fault diagnosis of machines
AU - Zhang, Yongchao
AU - Ren, Zhaohui
AU - Feng, Ke
AU - Yu, Kun
AU - Beer, Michael
AU - Liu, Zheng
N1 - Funding Information: This work was partially supported by the China Scholarship Council (CSC) under Grant 202106080066 . The authors are grateful for the support.
PY - 2023/5/15
Y1 - 2023/5/15
N2 - Cross-domain machinery fault diagnosis aims to transfer enriched diagnosis knowledge from a labeled source domain to a new unlabeled target domain. Most existing methods assume that the prior information on the fault modes of the target domain is known in advance. However, in engineering practice, prior knowledge of fault modes is rare in a new domain, in which there may be only partial source fault modes or some new fault modes. Furthermore, up to the present, almost all existing cross-domain fault diagnosis methods require the labeled source data during the model training process, which restricts their deployment on certain devices with limited computing resources. To this end, we propose a universal source-free domain adaptation method that can handle cross-domain fault diagnosis scenarios without access to the source data and is free of explicit assumptions about the target fault modes. More specifically, we develop a convolutional network with a Transformer as the attention module to extract discriminative feature information from the source data and then send the model and parameters to the target domain. In target domain training, we first propose a supervised contrastive learning strategy based on source class prototypes, which utilizes high-confident predictions to achieve source-free domain alignment and class alignment. Then, we also introduce a threshold-based entropy max–min loss to further align known class samples in the target domain or reject target outlier samples as an unknown class. Furthermore, we introduce self-supervised learning to further learn feature representations of the target domain to reduce the previous misclassification. A series of experiments on two rotating machine datasets demonstrate the effectiveness and practicability of the proposed method.
AB - Cross-domain machinery fault diagnosis aims to transfer enriched diagnosis knowledge from a labeled source domain to a new unlabeled target domain. Most existing methods assume that the prior information on the fault modes of the target domain is known in advance. However, in engineering practice, prior knowledge of fault modes is rare in a new domain, in which there may be only partial source fault modes or some new fault modes. Furthermore, up to the present, almost all existing cross-domain fault diagnosis methods require the labeled source data during the model training process, which restricts their deployment on certain devices with limited computing resources. To this end, we propose a universal source-free domain adaptation method that can handle cross-domain fault diagnosis scenarios without access to the source data and is free of explicit assumptions about the target fault modes. More specifically, we develop a convolutional network with a Transformer as the attention module to extract discriminative feature information from the source data and then send the model and parameters to the target domain. In target domain training, we first propose a supervised contrastive learning strategy based on source class prototypes, which utilizes high-confident predictions to achieve source-free domain alignment and class alignment. Then, we also introduce a threshold-based entropy max–min loss to further align known class samples in the target domain or reject target outlier samples as an unknown class. Furthermore, we introduce self-supervised learning to further learn feature representations of the target domain to reduce the previous misclassification. A series of experiments on two rotating machine datasets demonstrate the effectiveness and practicability of the proposed method.
KW - Domain adaptation
KW - Fault diagnosis
KW - Machinery
KW - Source-free
KW - Supervised contrastive learning
UR - http://www.scopus.com/inward/record.url?scp=85149750273&partnerID=8YFLogxK
U2 - 10.1016/j.ymssp.2023.110159
DO - 10.1016/j.ymssp.2023.110159
M3 - Article
AN - SCOPUS:85149750273
VL - 191
JO - Mechanical Systems and Signal Processing
JF - Mechanical Systems and Signal Processing
SN - 0888-3270
M1 - 110159
ER -